agentleFS
jpoindexterjpoindexter/design-case-study-skills/skills/portfolio/SKILL.md

portfolio

Skill1 starsChanged 4 months agoMITView on GitHub

Plan, build, critique, update, or ship a professional design portfolio as one evidence-backed system. Use when the user invokes /portfolio; needs portfolio positioning, project inventory and selection, personal brand, case studies, homepage/about/contact copy, information architecture, visual direction, website implementation, accessibility, review, or publishing; or wants to turn scattered work into a coherent portfolio for a specific role, client, or audience.

23 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use risk, ask and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • OlayeeeyoSkill

    risk

    评估当前整体持仓的风险敞口。当用户输入 /risk 时触发。适用场景:"我的持仓有什么风险?" "期权 assignment 风险大吗?" "持仓集中度怎样?

    Also in 3 projects that use this

  • dubwoahSkill

    ask

    Answer questions about the codebase without generating or modifying any code. Use when the user wants to understand how something works, find where something is defined, or explore the project structure. Invoked with /ask.

    Also in 3 projects that use this

  • AnthropicSkill

    architecture

    Create or evaluate an architecture decision record (ADR). Use when choosing between technologies (e.g., Kafka vs SQS), documenting a design decision with trade-offs and consequences, reviewing a system design proposal, or designing a new component from requirements and constraints.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 18 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    No mode means build. If a portfolio or project already exists, inspect it before asking for information that can be recovered from the …

    line 28
  2. Distinctive:

    Treat these as delegated capabilities, not separate reports.

    line 47
  3. Distinctive:

    Use these sequences as required routes, not suggestions:

    line 51
  4. Very specific:

    Do not append ship to build.

    line 67
  5. Very specific:

    Use assets/portfolio-workbook.md when the user needs a persistent artifact across sessions.

    line 82
  6. Distinctive:

    never invent research, metrics, testimonials, awards, employers, responsibilities, or impact.

    line 93
  7. Very specific:

    Use [evidence needed: …] for material gaps.

    line 95

“Common” is measured: kept by at least 6 in 10 of the 18 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Portfolio Dispatcher

Build a portfolio around what the intended audience must understand and believe. Do not begin with a visual theme or force every project into the same story.

Resolve the mode

Interpret the first argument as a mode when present:

CallOutcome
/portfolio buildRun the full workflow from evidence inventory to validated portfolio
/portfolio positionDefine target audience, promise, differentiators, and proof requirements
/portfolio inventoryFind and classify usable projects, artifacts, metrics, and testimonials
/portfolio selectScore projects and choose the smallest credible set
/portfolio architectureCreate the sitemap, page hierarchy, labels, and content model
/portfolio case-studyRoute one project through /case-study
/portfolio writeDraft or edit homepage, work-index, about, resume, and contact copy
/portfolio designDefine or implement the portfolio’s visual and interaction system
/portfolio auditTest positioning, evidence, findability, writing, accessibility, and execution
/portfolio shipValidate, commit, push, and deploy only with explicit publishing authority
/portfolio updateRe-run inventory and replace stale or weak evidence without rebuilding everything

No mode means build. If a portfolio or project already exists, inspect it before asking for information that can be recovered from the files.

Route the installed library

Load only the routes required for the active mode. If the use dispatcher is installed, run its inventory sweep against the active phase so new relevant skills can join later.

NeedRoute
Career/studio positioning, niche, expertise signalsdesignbiz
Personal brand, identity, positioning, touchpointsbrand
Product, system, or AI project storycase-study
Brand-identity project story or precedentbrand-case-studies with case-study
Sitemap, labels, project taxonomy, findabilityinformation-architecture-and-navigation
Interface direction, design system, layout, polishdesign
Section-level image direction or generated visualsimagegen-frontend-web
Voice and removal of AI writing patternshumanizer, after evidence and structure are stable
Accessibility and inclusive implementationaccessibility-and-inclusive-design
GitHub publicationgithub:yeet, only after an explicit ship request

Treat these as delegated capabilities, not separate reports. The portfolio owns the audience, project set, cross-page hierarchy, and proof system; companion skills own their specialized decisions.

Mode plans

Use these sequences as required routes, not suggestions:

ModeOrdered plan
buildposition → inventory → select → architecture → case-study for each selected project → write → design → audit → local handoff
positiondesignbiz for audience/expertise strategy → brand for personal-brand expression → proof requirements
inventoryInspect authorized files and sources → evidence/permission ledger → gap list
selectPermission gate → PROOF scoring → coverage/redundancy/depth check
architectureinformation-architecture-and-navigation → content model → sitemap → page briefs
case-studycase-study; add brand-case-studies for identity work
writePortfolio page copy → evidence check → humanizer voice pass
designdesign; add imagegen-frontend-web only when meaningful new imagery is needed
auditPositioning/evidence gate → rendered findability test → accessibility-and-inclusive-design → execution checks
shipaudit must pass → github:yeet → configured deployment path
updateInventory delta → replace or repair weak proof → rerun affected case-study/page/audit stages

Do not append ship to build. Run ship only when the user explicitly requests publishing.

Full workflow

  1. Target and position — name the role, client, or audience; their decision; desired next action; constraints; and proof requirements.
  2. Inventory — inspect projects, artifacts, metrics, research, screenshots, source files, testimonials, and permissions.
  3. Select — apply references/project-selection.md; choose the smallest set that proves the positioning without repetition.
  4. Architect — apply references/page-system.md; design the content model and sitemap before visual styling.
  5. Build stories — route each selected project through case-study; do not duplicate its internal framework here.
  6. Write connective copy — draft the homepage, work index, about page, project cards, resume bridge, and contact path.
  7. Design and implement — route the interface through design; generate new imagery only when it carries meaning.
  8. Validate — apply references/portfolio-quality-gate.md against the actual rendered portfolio.
  9. Hand off — leave a verified local artifact and one re-entry step.
  10. Ship only on request — publish only when the user explicitly invokes or requests the ship phase.

Use assets/portfolio-workbook.md when the user needs a persistent artifact across sessions.

Evidence contract

For every project and portfolio-level claim:

  • name the user, organization, or system affected;
  • state the designer’s actual ownership and collaborators;
  • connect artifacts to decisions, not decoration;
  • distinguish shipped outcomes from observations, reports, and proposals;
  • mark confidential, redacted, reconstructed, or unavailable evidence;
  • never invent research, metrics, testimonials, awards, employers, responsibilities, or impact.

Use [evidence needed: …] for material gaps. A visually strong page does not compensate for unsupported claims.

Output by default

Return or create:

  1. target-audience and positioning statement;
  2. evidence inventory;
  3. scored project shortlist;
  4. portfolio sitemap and page briefs;
  5. case-study production queue;
  6. portfolio-level copy;
  7. design/build direction;
  8. validation results and unresolved gaps;
  9. one next action when work remains.

Example

/portfolio build Inspect my project folders and build a portfolio for senior
AI product design roles. Prioritize systems thinking, agent control, and shipped work.

Expected outcome: the skill inventories evidence, selects non-redundant projects, defines positioning and sitemap, routes each selected project through /case-study, builds or directs the site through /design, and reports the executed validation path before calling it ready.

Done

The portfolio is ready only when the actual rendered artifact lets the target audience answer:

  • Who is this person, what kind of work do they do, and for whom?
  • Which projects best prove that claim?
  • What did they own, decide, and change?
  • What evidence supports the result?
  • Can I find the right depth quickly on desktop, mobile, keyboard, and assistive technology?
  • What should I do next, and does that action work?

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.